[unreadable] Neuroimaging and genomics are two primary evolving arenas of modern neuroscience. However, despite the attractiveness of combining these two approaches they have rarely been merged to fully capitalize on their power. These fields, while disparate, have common features such as an ability to obtain very precise, non-invasive quantitative information using complex computational modeling. Bridging quantitative neuroimaging based phenotyping and statistical genomics creates a new field - genomic imaging. The purpose of this application for a mentored career development award, is to transition Peter Kochunov from a neuroimaging specialist to an independent investigator in genomic imaging. This will combine neuroimaging and analytical genomics methodologies to execute studies that will contribute to a better understating of the genetic underpinnings of brain structures. This could provide the basis for discovering new molecular targets and other approaches for the characterization, treatment and potential prevention of brain disorders and will constitute a significant advance for basic neuroscience. Genetically complex brain diseases cost the United States as much as $1.2 trillion annually. Insight into the biological underpinnings that predispose individuals to these types of illnesses hold the promise of yielding new therapeutic strategies and a significant reduction of this considerable health and financial burden. The outcome of this training will result in genomic image processing methods by which one can perform statistical genomics measurements on multi-dimensional phenotypes describing inter-subject cerebral variability in a pedigree-type scenario. This project will use state-of-the-art 3D MR brain images obtained from baboon and human pedigrees and genetic maps to calculate genetic control of gyral gray matter thickness, cortical shapes and regional white-matter anisotropy. We will perform interspecies comparisons and study the genomics of age-related atrophy trends in humans. The data sample consists of 200 pedigreed baboons (Papio hamadryas) and ~1000 human subjects from a previously studied Mexican- American pedigree that are genotyped for a 10 cM genetic linkage map and imaged with a state-of-the-art MR structural imaging protocol. Novel variance components methods will be developed to allow genetic analysis of the MR-derived multi-dimensional phenotypes. In addition to generating valuable, new data, this project will allow Peter Kochunov to undertake rigorous training in analytical genomics and accumulate first hand experience in executing genomic studies. [unreadable] [unreadable] [unreadable]